The GEO Show

OpenAI Just Turned Websites Into AI Agent Tools

Welcome to Episode 10 of The GEO Show, where we break down the biggest developments in Generative Engine Optimization (GEO), AI search, and the rapidly changing world of SEO. In this episode, Paris Childress, founder of Hop AI and co-founder of GEOforge, runs through six stories reshaping AI search this week, starting with OpenAI's new WebMCP standard, which lets ChatGPT take structured actions directly on a website instead of clicking and scrolling like a human. In this episode: 🤖 OpenAI turns websites into agent-operable interfaces with WebMCP ChatGPT's desktop browser now supports site tools built on WebMCP, an experimental proposed standard that lets websites expose structured actions directly to AI agents instead of forcing them to navigate by clicking and typing. OpenAI is running a WebMCP Developer Challenge, and supported sites can let agents search, edit, and complete tasks directly. The web is acquiring a new optimization surface: not just can AI find you, but can AI use you, and agent readiness now belongs in every technical GEO and SEO audit. 📊 A 107-million-answer study says owned websites barely register as IT services citation sources Trendos analyzed 107 million AI answers across ChatGPT, Perplexity, Gemini, and Google AI Overviews. Among top citation sources for IT and solution services, 51% were community and UGC sources, 47% were independent editorial references, and just 2% were brand-owned websites. Frequently cited B2B sources included G2, Gartner, Clutch, GoodFirms, SourceForge, and Slashdot. For B2B technology, GEO is increasingly an off-site authority problem: the win often isn't publishing everything on your own site, it's pitching your information gain to the higher-authority publications already winning the citations. 📈 First-party Google data shows AI visibility is extremely concentrated A 99-day Search Console analysis of an established information site recorded 365,994 Google Generative AI impressions across 136 URLs, and just 14 of those pages generated roughly 81% of page-level AI impressions. It's the SEO world's 90/10 rule showing up again in GEO: a small minority of your pages will carry an outsized share of your AI visibility, and the job is to identify those pages and replicate what's working about them. 💬 Reddit dominates 83.9% of Google's discussions and forums results Ahrefs analyzed 145 million US Google results and found discussion and forum results appeared in 11.7% of them, with Reddit showing up in 83.9% of that feature globally, followed by Facebook and Quora at a distance. Longer, informational queries were especially likely to trigger discussion results. Reddit isn't just a big player in UGC citations, it functions like the Google of the forum landscape itself. 🎯 ChatGPT sometimes targets individual subreddits before retrieval A captured ChatGPT search explicitly targeted a subreddit, retrieved 48 Reddit threads among 71 results, and gave Reddit 6 of 8 citations. Four days earlier, another query retrieved 84 Reddit pages and cited none of them. Reddit's recent aggregate citation decline doesn't mean Reddit is off: source selection depends heavily on the underlying buyer question, so track citation share within your own category rather than assuming a blanket trend. 🏢 Frost & Sullivan formally enters the GEO advisory market Frost & Sullivan launched an AI brand equity and market presence advisory service using Optivara's GEO platform, benchmarking brands against competitors, analyzing how AI represents them through the funnel, and adding analyst interpretation on top. Raw share-of-voice monitoring is becoming table stakes rather than a premium feature. Real differentiation now comes from vertical expertise, proprietary brand knowledge, implementation, and above all, pipeline impact: asking customers how they heard about you and tracking that back to the sales pipeline. Subscribe to The GEO Show for ongoing analysis of the news, experiments, strategies, and emerging tactics shaping visibility across ChatGPT, Google AI, Perplexity, Claude, and the rest of the AI search ecosystem. 💬 Question: If you asked your last ten leads "how did you hear about us," how many would say an AI chatbot, and would you even know?

• Transcript

97% of Websites Are Invisible to AI Agents

Welcome to Episode 9 of The GEO Show, where we break down the biggest developments in Generative Engine Optimization (GEO), AI search, and the rapidly changing world of SEO. In this episode, Paris Childress, founder of Hop AI and co-founder of GEOforge, runs through seven stories reshaping AI search this week, starting with a new BrightEdge product that found 97% of the top 20,000 websites it analyzed aren't built for AI agents at all. In this episode: 🌍 Global GEO performance is a misleading KPI Profound's Summer 2026 Index analyzed 19.6 billion real user conversations and found that in 80% of industries with cross-market data, at least one European market had a number-one brand that had never ranked first in the US. Country-specific domains accounted for 46-59% of citations where a new regional leader emerged. International SaaS and cybersecurity companies should build country- and language-specific prompt sets, earned media strategies, and citation benchmarks per market, rather than assuming US visibility travels. 🛠️ BrightEdge launches an agent-optimization layer at the CDN level BrightEdge's new Agent Edge product detects AI agent requests and serves reduced, machine-oriented content instead of a full human-facing page. BrightEdge found that 97% of the top 20,000 websites it analyzed weren't serving any optimized machine format. Early customer Arm cut its payload size by about 90%, while Bloomfire reported a 30% lift in direct AI referrals. The market is moving from "can the model cite us" to "can the agent efficiently interact with us," and that shift is coming for e-commerce first, where agents will soon be shopping on behalf of their users. 🎥 A creative agency partners directly with an AI visibility platform on creator content Profound partnered with creative agency Influencer on August 26 to connect creator content and conversations directly with AI recommendation visibility. Creator marketing, digital PR, and GEO are converging into one authority-building discipline. For B2B brands pursuing GEO, that means mapping the YouTubers, LinkedIn experts, podcasters, and analysts AI engines are already referencing and integrating them into earned-authority campaigns. 📚 Listicles are down, but nowhere near dead Kevin Indig's analysis of 60,000 US Google queries and 5.32 million results, published on Search Engine Land, found listicles still appear in the top 10 for about 55% of queries, even as their share of the number-one position fell roughly 4 percentage points since January. 62% of vendor-authored listicles gained organic traffic, with a median gain of 38%. Google appears to be pruning weaker listicles, not banning the format: keep comparison content only where you can back it with real methodology, testing, and differentiated evidence. 📈 Top-of-funnel content is holding up in cybersecurity and SaaS A Search Engine Land analysis comparing traffic across 30 large cybersecurity and SaaS publishers found every sampled domain improved its top-of-funnel traffic year over year, while finance, healthcare, legal, and consumer tech publishers all declined. AI doesn't eliminate informational content equally across categories; complex, fast-changing B2B subjects still reward research depth that one generated answer can't fully satisfy. 🤖 Claude and Claude Code behave like two different search engines Profound's research, covering between roughly 24,000 and 135,000 responses, found that with web search enabled, Claude searched the web in 93% of its responses versus just 13% for Claude Code, and the two products overlap on only 20% of mentioned brands. Nearly three-quarters of Claude Code's agent traffic went to documentation, informational, and pricing pages, versus just 5% for Claude. Developer-focused SaaS and cybersecurity products need a rich documentation layer with precise, machine-retrievable compatibility, pricing, API, and deployment facts. ⚠️ Did Google's August spam update target AI-generated content? Search Engine Journal documented multiple reports of mass-generated AI sites losing visibility during Google's August spam update, but Google hasn't confirmed the update specifically targeted AI content, and its policy remains technology-neutral: AI content is allowed, but creating many low-value pages primarily to manipulate rankings is scaled content abuse. The risk was never using AI. It's industrialized commodity publishing with no real information value behind it. Subscribe to The GEO Show for ongoing analysis of the news, experiments, strategies, and emerging tactics shaping visibility across ChatGPT, Google AI, Perplexity, Claude, and the rest of the AI search ecosystem. 💬 Question: If an AI agent showed up at your site right now, could it actually get anything done, or would it just get a bloated page it can barely read?

• Transcript

The B2B Buyer Journey Has Already Collapsed

Welcome to Episode 8 of The GEO Show, where we break down the biggest developments in Generative Engine Optimization (GEO), AI search, and the rapidly changing world of SEO. In this episode, Paris Childress, founder of Hop AI and co-founder of GEOforge, runs through ten stories reshaping AI search this week, starting with new traffic data showing the SaaS buyer journey has collapsed from multiple searches into a single, long-threaded AI conversation. In this episode: 📉 SaaS organic traffic is down, but AI-referred visitors convert far better First Page Sage analyzed 3.4 billion sessions across 218 client sites and found B2B SaaS organic sessions down 29% to 29.4% from baseline, while AI platforms now drive 11.8% of SaaS sessions. ChatGPT referrals convert to leads at 4.7%, versus 1.9% for Google organic, roughly a 3x lift. Separately, Brainlabs' data across 54 advertisers found organic sessions down 10.5%, AI referrals up 163%, and AI-driven key events up 335%. The buyer journey hasn't shrunk. It's collapsed into one educated, ready-to-convert conversation with an LLM. 🔎 Keenable raises $26 million to build a search engine for AI agents Keenable emerged from stealth on August 25 with a $26 million seed round led by Accel, founded by former Yandex search AI executives. It says it has indexed more than 100 billion documents and is already serving AI labs and inference providers. The retrieval layer beneath AI assistants is quietly becoming its own infrastructure market, and Google and Bing may not stay the only gateways to the web for agents. 📰 Featured's citation data rewrites the off-site GEO playbook Featured analyzed 22,881 Perplexity citations across 11,499 domains and found 34.5% came from domains with Moz domain authority below 40, while editorial sources made up 64% of citations versus just 1.7% for brand-owned content. Off-site GEO is looking more like citation-informed digital PR than conventional high-authority link building: build outreach lists from the publications that actually appear in buyer prompt citations, not from generic DR or DA thresholds. 🛠️ Webflow brings site audits directly into Codex and ChatGPT Webflow's August 24 integration extends its MCP-powered capabilities into Codex and ChatGPT, covering SEO and AI-answer audits, accessibility checks, broken link detection, CMS management, and safe publishing. Routine audit work is increasingly automated through MCP connectors. What's left as the premium layer is strategy, proprietary knowledge, experimentation, and knowing which changes actually move the needle commercially. 📱 BrowserAct launches free, browser-based GEO reporting BrowserAct's free report, launched August 25 for ChatGPT, Claude, Gemini, and Perplexity, executes questions inside the actual consumer web interface rather than through an API, preserving the prompt, the visible answer, citations, and screenshots. Almost every other AI visibility tracker is API-based, so this browser-first measurement approach is worth watching for how its results compare. 💰 Buyers are funding GEO, but mostly still call it SEO Fractal surveyed 343 US marketing decision-makers: 81% still describe AI search visibility internally as SEO, while marketers report allocating an average 24% of search and content budgets to AI visibility, and two-thirds have already used ChatGPT, Gemini, or Perplexity to evaluate marketing vendors. The budget is moving faster than the terminology, so pitches that lead with the combined SEO-plus-AI-visibility problem still land best. 📊 New data shows how differently each AI engine cites evidence PromptScout's dataset of 5,436 completed answers found median source counts per answer of 38 for ChatGPT, 15 for Perplexity, 7 for AI Overviews, 6 for Gemini, and just 3 for Bing Copilot. A specialist page has a realistic shot at a ChatGPT citation and a much longer shot at Google AI Overviews, so the right citation-building strategy is genuinely different model by model. 🪞 A GEO case study exposes the gap between being cited and being known Seek Labs reports one site reaching 451 citations across 310 pages, with citations up 82.6% while brand mentions rose only 10%, a roughly 8.5-to-1 ratio. AI can repeatedly use a company's content while barely strengthening its brand presence, which means citation share, brand mentions, and recommendation sentiment all need to be tracked as separate outcomes. 🏚️ Some GEO losses trace back to real product problems, not visibility Search Engine Land reports cases where sophisticated brands were omitted from AI recommendations because the available evidence surfaced real issues: missing integrations, support problems, reliability concerns. GEO is becoming a customer and product intelligence system too, and those findings need to route to product, customer success, or leadership, not stay contained in marketing. ⚠️ Brand accuracy needs to become a GEO KPI An Imperva rank audit tested 1,257 factual claims across 182 businesses recommended by a search-enabled OpenAI configuration. Nearly 2% of claims directly contradicted available evidence, 11% of businesses had at least one incorrect claim, and another 38% of claims were unsupported rather than proven false. Discoverability isn't the whole game. What AI says about your certifications, pricing, and capabilities also has to be correct. Subscribe to The GEO Show for ongoing analysis of the news, experiments, strategies, and emerging tactics shaping visibility across ChatGPT, Google AI, Perplexity, Claude, and the rest of the AI search ecosystem. 💬 Question: If your GEO reporting still leads with referral traffic, what would your numbers look like if you measured the funnel instead?

• Transcript

Google AI Mode Loses 62% of Its Citations Every Day

Welcome to Episode 7 of The GEO Show, where we break down the biggest developments in Generative Engine Optimization (GEO), AI search, and the rapidly changing world of SEO. In this episode, Paris Childress, founder of Hop AI and co-founder of GEOforge, runs through seven stories reshaping AI search this week, including new daily-tracking data showing just how volatile AI citations really are from one day to the next. In this episode: 🔍 Google says there's no special technical GEO layer for AI Search John Mueller confirmed there's nothing website owners need to do specifically for Google's AI Overviews or AI Mode beyond strong SEO fundamentals: crawlability, indexation, clear information architecture, and entity clarity. It's a Google-only position and doesn't mean AI visibility equals rankings, but it does undercut the mysterious "Google GEO hacks" some vendors are selling. 🧬 Pew Research finds AI fingerprints on 10% of the entire web, 35% of pages published since ChatGPT Pew analyzed 490,000 English-language pages from Common Crawl's July 2026 sample. AI authorship or editing shows up in 10% of all pages, and in 35% of pages with a publication date after ChatGPT launched in November 2022. AI-generated content isn't a differentiator anymore. It's ambient infrastructure, and the only moat left is information gain: subject matter expert interviews, proprietary data, benchmarks, and firsthand experience. 💻 NVIDIA and Perplexity launch Portable Computer for local-first AI Following earlier rumors of an NVIDIA investment valuing Perplexity at $30 billion, the two companies have introduced Portable Computer, which runs Perplexity entirely on-device and escalates to the cloud only when a task requires it. It's a serious signal for privacy-conscious companies not yet ready to send proprietary data into cloud-hosted models, and an early look at what AI on the edge looks like in practice. 📊 The first public, searchable ChatGPT Ads library has launched LLMPulse launched an independent ChatGPT Ads library on August 21 and updated it again on August 24, now covering more than 72,000 observed ads across nine countries, searchable by advertiser, message, country, and date. It's sampled, observational data, not official OpenAI numbers, but it's the first real crack at competitive intelligence for conversational ads. 📉 Daily AI citations are far noisier than classic SEO rankings Vendor research from Chloro tracked identical prompts across consecutive daily runs from August 1 to 16. A previously cited domain stayed cited the next day just 37.8% of the time in Google AI Mode, 49.9% in ChatGPT, and 68.5% in Perplexity. Losing a citation today is statistical noise, not proof your optimization failed. Report GEO from thirty to forty runs at a time, not a single execution. 📰 Muck Rack's 25-million-link dataset shows AI source strategy has to be platform-specific Muck Rack's analysis of more than 25 million cited links across ChatGPT, Gemini, and Claude found citation behavior varies materially by platform and query type. Industry trend queries cite journalism at more than twice the rate of how-to queries, and more than half of journalism citations came from content published in the last twelve months. Domain authority doesn't factor in at all. Relevance does. 🤖 Agent experience is emerging as the discipline downstream of GEO RedMonk's new framework draws a clear line between GEO, which gets a product discovered or recommended, and agent experience (AX), which determines whether an agent can actually use it. Agents are already crawling sites at a far higher rate than human visitors. The next question isn't just can AI recommend you, it's can AI operate you. Subscribe to The GEO Show for ongoing analysis of the news, experiments, strategies, and emerging tactics shaping visibility across ChatGPT, Google AI, Perplexity, Claude, and the rest of the AI search ecosystem. 💬 Question: If you reported a lost AI citation as a loss this week, would it still look like a loss after forty runs instead of one?

• Transcript

GEO Deep Dive: Building Citations Through Outreach

Welcome to a bonus, live session of The GEO Show. This one's a conversation, not a solo digest: Paris Childress, founder of Hop AI and co-founder of GEOforge, sits down with Simon Young, founder of Question.Marketing, to work through where AI citations are actually coming from and what to do about it. In this episode: 🤖 A fake bot impersonating OpenAI's crawler gets caught in the wild Hop AI detected a bot on a client's Cloudflare setup using the exact same name as OpenAI's real crawler, but from a different IP address. Impersonating a known crawler name is trivial, and it's turning bot policing and selective blacklisting into a real part of the technical side of GEO. 📊 The number: brand-owned sites get just 2% of AI citations A study analyzing 107 million AI answers across ChatGPT, Perplexity, Gemini, and Google AI Overviews in the IT services vertical found 51% of citations came from community/UGC sources, 47% from independent editorial sites, and only 2% from brand-owned websites. Publishing on your own domain and hoping AI notices isn't a citation strategy on its own. ✉️ Why "add us to your listicle" outreach caps out at a 1-2% reply rate Asking an already-published site to edit in a new brand rarely works, most people don't want to revisit something they've already shipped. The pivot: pitch fresh domains that haven't already covered the topic, and offer them something with real value instead of a favor. 🎁 Information gain becomes the outreach currency Instead of a reciprocal link swap, give a target site an exclusive, high-information-gain piece of original research for free. They get a story with real substance. You get cited as the source, even without a backlink, in a strategy that reads a lot like classic PR pitching. 🔗 Mentions now beat backlinks for AI visibility Answering a live audience question from Ashley Pearson, Paris and Simon agree that AI doesn't check domain authority or a dofollow attribute the way search crawlers once did. With a similar success rate, prioritizing brand-mention requests over backlink requests is the better bet, since mentions convert at roughly 5% versus roughly 1% for link asks. 🔀 What's next for ContentForge and SiteForge Hop AI is rethinking how much of its best content should get pitched externally, as exclusive research for third-party sites, before it ever touches the brand's own domain. Subscribe to The GEO Show for ongoing analysis of the news, experiments, strategies, and emerging tactics shaping visibility across ChatGPT, Google AI, Perplexity, Claude, and the rest of the AI search ecosystem. 💬 Question: If you had one piece of genuinely original research today, would you publish it on your own site first, or give it away exclusively to build a citation somewhere with more authority?

• Transcript

GEO Tools Have Their Own AI Visibility Problem

Welcome to Episode 6 of The GEO Show, where we break down the biggest developments in Generative Engine Optimization (GEO), AI search, and the rapidly changing world of SEO. In this episode, Paris Childress, founder of Hop AI and co-founder of GEOforge, runs through six stories reshaping AI search this week, including new data showing the GEO tools category can't get itself cited by the very engines it's built to track. In this episode: 🧠 ChatGPT search is explicitly personalized before retrieval OpenAI's updated Help Center documentation confirms ChatGPT can rewrite one prompt into multiple searches (query fan-out) and use saved memory of a user's conversation history to construct those queries. Two people asking the same question can now trigger very different retrieval paths before the model ever sees a source, which means GEO testing needs buyer personas and contextual variants, not one canonical prompt. 💡 Information gain is becoming the clearest antidote to AI content sameness Ahrefs' August 20 Deep Dive argues useful content has to contribute original data, first-hand experience, or genuinely new perspective beyond what already exists, though Google hasn't confirmed its patented information gain score as a live ranking factor. It's the core reasoning behind GEOforge's own BaseForge approach: ground content in proprietary knowledge so it earns citations instead of reading as derivative AI output. 🎯 Topical depth beats broad AI visibility for getting a brand actually named Semrush analyzed 283,215 citation observations across 1,094 ChatGPT categories and found brands were cited and named in 34% of appearances within closely related categories, versus just 9% in distant ones. Depth inside two or three strategic categories now correlates more strongly with brand mentions than spreading thin across many. 🏛️ GEO has officially entered institutional procurement South Dakota State University's public LLM search optimization RFP carries an $80,000 annual ceiling, a two-year initial term, and roughly 9,000 pages in scope, requiring AI answer monitoring, Drupal integration, audit logging, US data residency, and SOC 2 Type II or ISO 27001 credentials. GEO is moving from experimental marketing spend into governed enterprise procurement. 👻 The GEO tools category has an AI visibility problem of its own SEO Rocket measured 6,322 citations across 56 SEO/GEO tools using Ahrefs Brand Radar. Established SEO platforms averaged 131 cited AI answers each; dedicated GEO tools averaged just 5, with Perplexity generating more than a third of all sampled citations. Merely entering the GEO category, GEOforge included, doesn't make a brand part of AI's category model — every tool has to prove its own AI visibility to earn its place in it. 🕷️ AI crawler access is becoming a governance and monetization decision Cloudflare's AI Crawl Control now offers crawler-level monitoring, allow/block policies, robots.txt compliance tracking, and a private-beta pay-per-crawl capability, though WAF rules can override crawler policies. With spoofed bots now posing as legitimate crawlers like OpenAI's, brands need to actively analyze crawl logs rather than treat AI bot access as a simple yes/no decision. Subscribe to The GEO Show for ongoing analysis of the news, experiments, strategies, and emerging tactics shaping visibility across ChatGPT, Google AI, Perplexity, Claude, and the rest of the AI search ecosystem. 💬 Question: If your brand just entered the GEO category, are you actually measuring whether AI cites you in it, or just assuming membership is enough?

• Transcript

GPTBot Found Zero Pages Behind Your JavaScript Links

Welcome back to The GEO Show, episode five. Every week we cut through the noise around AI search and Generative Engine Optimization (GEO) and tell you what's actually worth your attention. In this episode, Paris Childress, founder of Hop AI and co-founder of GEO Forge, runs through eleven stories from the past week in AI search, ad platforms, and GEO strategy. In this episode: 📊 ChatGPT Ads adds view-through conversion tracking OpenAI now reports conversions when someone sees but doesn't click an ad within a one-day window. That's going to muddy attribution further when the same lead is claimed as last-click by Google Ads and view-through by Meta, LinkedIn, and now ChatGPT Ads too. Add "how did you hear about us" to your lead forms with an AI chatbot option. 💼 GEO becomes an in-house hire Cisco Careers posted a job ad for an "SEO/GEO Content Strategist." Replit is recruiting a lead role spanning GEO and SEO, tied to sign-ups and revenue. Large enterprises are now insourcing GEO the same way they once insourced SEO. 🕸️ The JavaScript crawler blind spot A 41-day test across roughly 1,000 pages logged 30,180 bot requests. Googlebot crawled JavaScript-injected links fine. GPTBot, ClaudeBot, Perplexity, and other AI crawlers found zero of those pages. After converting the links to plain HTML, GPTBot discovered 250 pages within 48 hours. ⚙️ A new entrant merges monitoring and execution Surva.ai now spans ChatGPT, Perplexity, Claude, Gemini, and Google AI Overviews with visibility tracking, competitor analysis, and content generation in one platform. The fully integrated monitor-diagnose-draft workflow is becoming commoditized fast. 🔍 AI search runs on tasks, not keywords Google's own AI Mode data shows the average query is three times longer than a traditional search. Planning-related queries have grown 80% faster than AI Mode queries overall. Content briefs need to model the whole journey, not one keyword. 🎯 The real GEO opportunity may be smaller than the keyword universe Duane Forrester of Search Engine Journal argues AI systems collapse many phrasings into small consideration sets. GEO strategy may end up looking more like category ownership than long-tail keyword capture. He discloses a commercial interest through Citation IQ, so treat this as a hypothesis. 🧰 Interactive tools over-index on AI referral traffic StudioHawk analyzed 1.2 million AI referral sessions across 600+ businesses. Interactive tools, templates, and calculators earned 7.5x their fair share of AI traffic. Standard blog posts under-indexed. 📍 Gemini's citations are wildly unstable Steady Demand analyzed 14,472 citations from 1,487 local queries across 50 US metros. Repeated identical queries shared citation sources only about 40% of the time. Gemini and ChatGPT cited the same domains just 8% of the time. 📉 Search Console data can look great and mean nothing Ivris Tech reports two pages with 542,615 impressions at an average position of 3.6, and only seven clicks over five months. The authors can't rule out bot and AI agent traffic inflating those impressions. 🧪 GEO audits are commoditizing fast Romanian startup Human AI Labs launched Blindspot, a 70-criteria AI readiness audit. Beta audits averaged 48 out of 100. The scores haven't been shown to cause higher AI visibility, the same pattern SEO audits went through years ago. ⚖️ EU AI Act transparency rules are now live Article 50 has applied since August 2nd, 2026. Providers face machine-readable marking requirements, and penalties can reach €15 million or 3% of worldwide annual turnover. Subscribe so you don't miss next week's episode, and follow along at getgeoforge.com and hoponline.ai. 💬 Question: If Google can index your JavaScript pages but GPTBot can't read them, does that page even count as "ranking" anymore?

• Transcript

AI Citations No Longer Need a Top-10 Ranking

Welcome to Episode 4 of The GEO Show, where we break down the biggest developments in Generative Engine Optimization (GEO), AI search, and the rapidly changing world of SEO. In this episode, Paris Childress, founder of Hop AI and co-founder of GEOforge, runs through eight stories reshaping AI search this week, including new data on why ranking #1 on Google no longer guarantees you an AI citation. In this episode: 🔘 Google launches a user-controlled "Preferred Sources" button Google's new embeddable button, live since August 20, lets readers select a site as a preferred source, making it easier for that site to surface in Top Stories, AI Overviews, and AI Mode. More than 600,000 sources have already been selected. Adding this button to your blog is close to a no-brainer. 🆓 Llumo undercuts the AI-visibility tracking market with a free tier Llumo launched August 20 with free AI visibility software using a bring-your-own-API-key model, tracking mentions, competitors, citations, prompt responses, query fan-out, sentiment, and share of voice across ChatGPT, Gemini, Perplexity, Copilot, AI Mode, and AI Overviews. Basic prompt and citation monitoring is now a commodity; differentiation has to come from what's built on top of it. 🎥 Creator marketing budgets start chasing AI citations Digiday reports Zoom and agencies including Trevant and Crispin are auditing which creators get LLM citations and using that data to choose who they work with. GEO is now influencing influencer selection, not just distribution. 📱 Google AI Overviews pull social posts at real scale A BrightEdge study of 300+ million monthly US searches found Facebook cited 19.5 million times in AI Overviews, Instagram roughly 877,000 times, and TikTok 78,000 times. Exact answer relevance is beating follower size, and long-tail social posts are becoming part of the retrieval layer, not just a distribution channel. 📋 Your own "best tools" listicle can promote your competitor Search Engine Journal and Ahrefs tracked 9,886 AI answers across 34 self-promotional listicles. AI systems frequently used the list as a source without recommending its publisher, and in one case a competing conference got recommended in 43% of answers pulling from Ahrefs' own listicle. Listicles are also declining as a share of total citations. 📊 A blended AI visibility score is nearly meaningless Fractal ran 96 prompts 15 times each across ChatGPT, Gemini, and Claude, generating 4,320 responses and 8,500+ brand references. Only 11% of brands appeared across all three models; 77% appeared in only one. Report AI visibility by platform, not as one blended number. 🕷️ European sites are paying a much higher AI scraping tax Tollbit analyzed AI bots from 40 vendors across 3,906 publishers and found European sites getting scraped 4x more than North American sites, with the scrape-to-referral ratio worsening from 150:1 in Q1 to 227:1 in Q2. Crawler volume is a poor proxy for AI-referred traffic, and it's getting worse. 🎯 Ranking #1 on Google no longer guarantees an AI citation Ahrefs analyzed 1.4 million ChatGPT prompts and found only 38% of AI Overview-cited URLs ranked top 10 for their original query, down from 76% a year ago. You still need to be indexed and competitively ranked, but query fan-out now goes well past page one, so expanding topic coverage matters more than chasing the #1 spot. Subscribe to The GEO Show for ongoing analysis of the news, experiments, strategies, and emerging tactics shaping visibility across ChatGPT, Google AI, Perplexity, Claude, and the rest of the AI search ecosystem. 💬 Question: If you're still reporting rank position as your main AI-visibility metric, what would you actually see if you measured citation rate instead?

• Transcript

Reddit's Collapse, Google's New UI, and Zero-Click Everything

Welcome to Episode 3 of The GEO Show, where we break down the biggest developments in Generative Engine Optimization (GEO), AI search, and the rapidly changing world of SEO. In this episode, Paris Childress, founder of Hop AI and co-founder of GEO Forge, runs through nine stories reshaping AI search this week, including one citation collapse nobody has explained yet. In this episode: 🔍 ChatGPT is doing a lot more research for B2B software queries Nective tested 4,000 prompts and found average ChatGPT fan-out queries rose from 2.17 to 7.61, with software queries alone averaging 10.7 searches per prompt. B2B recommendations are becoming a research workflow inside LLMs, not a single query. 📊 Your GA4 numbers may be lying about AI Overviews traffic A nine-month dataset covering 51,200 tracked events found that 22.4% of detected AI Overview events were attributed to direct traffic instead of organic. In a zero-click world, don't take Search Console or GA4 at face value. 🧠 Google's own research: LLMs know facts but still can't retrieve them Gemini 3 Pro and ChatGPT-5 encoded 95-90% of tested facts correctly, but failed to directly recall 26-34% of those same facts. Training-data accuracy and real-time discoverability are two separate problems, and you need to win at both. 🛠️ Google AI Overviews can now generate interactive interfaces, not just answers Search Engine Journal reports Google has started rolling out Generative UI, including custom layouts, simulations, and interactive tools. If your lead-gen strategy depends on a gated calculator, Google's AI answer can now reproduce it with no click required. 📣 ChatGPT Ads landed in Europe today OpenAI is expanding into 31 European markets with conversion optimization, geo-targeting, custom audiences, Pixel, and Conversion API measurement, the full performance-marketing stack. 📉 Reddit's share of ChatGPT citations collapsed 86% in three days According to Promptwatch, Reddit's average citation share fell from 3.83% to 0.52% between August 14 and 17. Reddit hasn't ruled out a collection issue. Either way, a citation strategy concentrated in one off-site channel just got a lot riskier. 📹 YouTube citations are now measurable at the creator level Profound shipped a citation breakdown that identifies cited channels, individual videos, and video categories. YouTube remains the number one citation source across Google's AI properties, and it's time to measure it like owned media. ⚠️ Even "stable" brand content is feeding LLMs bad facts Profound analyzed 158,000+ evaluated AI claims and found engine-level inaccuracy rates of 4.3-6.8%. Pricing and billing errors made up 24% of inaccurate claims despite being just 12% of claims evaluated, so keep your pricing current and visible. 💵 Commercial intent inside ChatGPT is rising fast Profound classified a year-long sample of 7.5 million conversations and found commercial intent share rose from 13.9% to 19.2% between June 2025 and June 2026, roughly 533 million weekly commercial conversations. It's still early, and that's the opportunity. Subscribe to The GEO Show for ongoing analysis of the news, experiments, strategies, and emerging tactics shaping visibility across ChatGPT, Google AI, Perplexity, Claude, and the rest of the AI search ecosystem. 💬 Question: If a channel you'd built your whole citation strategy around collapsed overnight, would you even notice before it showed up in your leads?

• Transcript

What It Actually Looks Like to Build AI Visibility for a Startup

Most brands have no idea whether AI recommends them. This is what the first day of fixing that actually looks like, with real numbers and nothing hidden. Paris Childress of Hop AI and Simon Young of Question.Marketing take one of Simon's clients, WhatsApp Business Platform Stitch AI, and start building its visibility inside ChatGPT, Google AI Overviews and Google AI Mode in public. No case study gloss. Day one, live, including the baseline nobody wants to publish: Stitch is mentioned in one prompt out of thirty. Over the next 60 days this series follows the whole build. This first session covers how the prompt set gets chosen, how share of voice is measured and why it is harder than SEO ever was, how a knowledge base of 1,335 sales call recordings gets turned into content AI models will cite, and what "information gain" actually means when everyone is drowning in AI slop. WHAT YOU'LL TAKE AWAY How to attribute leads from AI when there is no click to track Why the prompt set is the part most businesses get wrong How to find the words your customers really use, using competitor reviews Why first-party knowledge beats volume when you want to be cited How to baseline your share of voice before you spend a penny on content CHAPTERS 00:00 Who's on this and what we're building 01:00 From black hat SEO to answer engines: Simon's route in 02:00 Where did all the clicks go? 03:00 Calling out AI referral numbers nobody can prove 04:00 Is GEO a real channel, or SEO with a new sticker? 05:00 Zero-click search: why an AI referral leaves no trail 07:00 The "how did you hear about us?" fix for AI attribution 08:00 What Stitch AI does, and why the category is hard to name 10:00 "Bulk WhatsApp": the term customers search, not the one you'd pick 13:00 Using G2 to find the category your competitors sit in 14:00 Mining 500 competitor reviews for the voice of the customer 15:00 Inside the knowledge base: 1,335 sales calls in 24 hours 17:00 Information gain explained: scoring what models haven't seen 18:00 AI slop, the Anthropic watermark and where LinkedIn lands 21:00 Does AI-generated content get penalised? The honest answer 23:00 Building 30 prompts across top, middle and bottom of funnel 25:00 Can you retire a prompt once you own it? 26:00 The baseline: mentioned in 1 prompt out of 30 27:00 Which engines we measure, and which we ignore for now 29:00 Turning the gaps into a content plan 31:00 Case studies, comparison pages and keeping data safe 32:00 The 60-day goal and what happens on Day 2 ABOUT THE GUESTS Paris Childress is the founder of Hop AI and GEOForge, the platform used throughout this session to build the knowledge base, track prompts and measure share of voice. https://hoponline.ai Simon Young runs QuestionMarketing, a done-for-you AI implementation and answer engine optimisation consultancy in Doncaster, South Yorkshire. He has been publishing on answer engine optimisation since October 2019, more than three years before ChatGPT existed. https://question.marketing https://www.linkedin.com/in/aeo Stitch AI is an official Meta business partner and WhatsApp communication platform for businesses running messaging at scale. SUBSCRIBE for Day 2, where the first content drafts come out of the knowledge base and we see whether the share of voice moves. Questions for the next session go in the comments. We read them live. Generative engine optimization (GEO), also called answer engine optimisation (AEO) or AI search optimisation, is the practice of getting a brand named and cited inside AI-generated answers rather than ranked in a list of links. This session covers AI visibility, share of voice measurement, LLM citations, prompt tracking, ChatGPT, Google AI Overviews, Google AI Mode, zero-click search, information gain, first-party knowledge bases and build-in-public GEO strategy for startups and B2B SaaS. #GEO #AEO #AIVisibility #GenerativeEngineOptimization #AISearch #ChatGPT #B2BSaaS

The Importance of a Knowledge Base in GEO

Most companies are approaching Generative Engine Optimization (GEO) from the wrong direction. They are asking: “How can we use AI to produce more content?” The better question is: “What proprietary knowledge do we have that AI cannot find anywhere else?” In the second episode of GEO Day 1, Simon Young and Paris Childress will examine why a proprietary knowledge base is becoming the most important content asset a company can build. The internet is rapidly filling with traceable AI slop: articles assembled from the same public sources, repeating the same claims, following the same structures and offering no meaningful new information. The wording may be different, but the underlying content is not. What's more... Anthropic just announced an AI watermark similar to Gemini's SynthID, and other LLMs are sure to follow suit. Soon, All AI-assisted content will be easy to spot everywhere it's published. This creates a fundamental problem. If your AI tools are grounded in the same public information available to everyone else, they will produce variations of what already exists. Publishing faster will simply help you create more interchangeable content. That content will struggle to earn attention, trust, links and citations—whether the reader is a human, Google or a large language model (LLM). The alternative is high information gain content. High information gain content gives the reader something genuinely new: an expert observation, original data, a customer insight, a tested methodology, a detailed case study, a contrarian conclusion or evidence that advances the existing conversation. But information gain cannot be manufactured through prompting alone. It must be grounded in proprietary knowledge. That means building a structured knowledge base containing the information your business knows but the wider internet does not: - Interviews with subject-matter experts - Customer questions, objections and buying language - Original research and internal data - Case studies with specific evidence and outcomes - Product knowledge and implementation experience - Proprietary frameworks and methodologies - Opinions formed through real-world experience - Lessons from successes, failures and edge cases This knowledge base becomes the grounding layer for content production. AI can then help retrieve, organize and transform that knowledge into useful assets without replacing it with generic internet consensus. The result is not “AI-generated content.” It is company-generated knowledge, structured and amplified by AI. In this LinkedIn Live session, we will discuss: - Why generic AI content is becoming increasingly easy to recognize - What “information gain” actually means in practice - Why prompting cannot compensate for weak source material - What belongs inside a proprietary knowledge base - How expert interviews can systematically capture institutional knowledge - How one source of proprietary insight can support articles, landing pages, sales content and other formats - How a knowledge base strengthens both traditional content marketing and GEO - Where human expertise must remain in the workflow - How startups can begin building this asset without a massive content operation This is not a session about producing more content. It is about building a source of truth that allows your company to publish content competitors cannot easily reproduce—and that humans and AI systems have a reason to trust, reference and cite. If AI can recreate your article without knowing anything about your company, the article probably contains very little defensible value. Join Simon and Paris for a practical discussion about building the knowledge foundation behind sustainable AI visibility.

Google's GEO Reporting Bug, Ahrefs' AI Volume Metric, and Apple's Crawler Surge

Welcome to Episode 2 of The GEO Show — where we break down the biggest developments in Generative Engine Optimization (GEO), AI search, and the rapidly changing world of SEO. In this episode, Paris Childress, founder of Hop AI and co-founder of GEO Forge, looks at what may be the beginnings of an entirely new search ecosystem — complete with its own analytics, advertising infrastructure, crawlers, and measurement challenges. In this episode: 📊 Google's new GEO reporting already has a data problem Google Search Console's new generative AI performance reporting is giving marketers first-party visibility into AI citations and impressions — but a logging bug is also a reminder that GEO measurement is still very immature. 💰 Advertising inside AI answers is becoming a real media channel Similarweb is now tracking advertising placements across ChatGPT, Google AI Mode, and AI Overviews. The bigger takeaway: paid AI search is quickly becoming measurable enough for marketers to treat it as a legitimate advertising channel. 🎯 OpenAI is building serious advertising attribution infrastructure ChatGPT advertisers now have access to increasingly familiar performance marketing technology, including pixel-based conversion matching. OpenAI's advertising stack is starting to look a lot more like the infrastructure marketers already know from Google and Meta. 🍎 Why is Applebot dramatically increasing its crawling capacity? Apple has reportedly expanded Applebot's crawl infrastructure by thousands of IP addresses. Apple hasn't explained why — but it's a fascinating signal about where Apple's AI and search ambitions could be heading. 🔎 Ahrefs is trying to estimate AI prompt demand Ahrefs has introduced an "AI-adjusted volume" metric designed to approximate demand across AI platforms. But can we ever have a true equivalent of SEO keyword volume for AI prompts? My prediction: probably not. 🏷️ AI has a brand categorization problem A large language model (LLM) can know plenty about your company while still failing to associate your brand with the category you want to own. That creates a critical new GEO challenge: category association. Comparisons, customer proof, partner references, reviews, earned media, and consistent positioning may become increasingly important signals for teaching AI systems where your brand belongs. 📉 “Ranking in ChatGPT” may be the wrong KPI entirely New research suggests different ChatGPT modes can rely on dramatically different retrieval systems and source sets. That's why GEO isn't simply about getting a webpage to “rank.” The real objective is getting the right passages, claims, and brand associations surfaced by AI systems at the right moment. And that requires marketers to rethink some of the fundamental assumptions we've carried over from SEO. Subscribe to The GEO Show for ongoing analysis of the news, experiments, strategies, and emerging tactics shaping visibility across ChatGPT, Google AI, Perplexity, Claude, and the rest of the AI search ecosystem. 💬 Question: Do you think we'll ever have a reliable equivalent of keyword search volume for AI prompts — or is prompt volume fundamentally impossible to measure?

• Transcript

ChatGPT Referral Traffic Jumps 300% as GEO Diverges From SEO

Welcome to Episode 1 of The GEO Show — a regular breakdown of the biggest developments in Generative Engine Optimization (GEO), AI search, and the changing world of SEO. In this episode, Paris Childress, founder of Hop AI and co-founder of GEO Forge, breaks down five stories that could have major implications for B2B and SaaS marketing teams. In this episode: 📈 ChatGPT referral traffic is surging Demandbase data shows ChatGPT-referred visits to B2B websites increased more than 300% year over year. Is GEO starting to become a true performance marketing channel? 🤖 AI chatbots are changing how B2B buyers choose software G2 research finds 71% of B2B software buyers now use AI chatbots for research — and 51% start with AI chatbots more often than Google. 🔎 OpenAI confirms “query fan-out” ChatGPT can turn a single prompt into multiple targeted searches. We look at why those hidden queries matter for GEO — and how marketers can use Bing Webmaster Tools to uncover and optimize against them. 📌 Pinterest shows what GEO at scale can look like Pinterest deployed a system using vision-language models, query prediction, dynamically generated collection pages, and internal linking — producing a reported 20% increase in organic traffic. ⚠️ Microsoft draws a line against “manipulative GEO” Bing is now explicitly warning that abusive GEO tactics can result in reduced visibility or even delisting. But Microsoft and Google continue to suggest that good GEO largely comes from good SEO. I’m not convinced. We also dig into one of the most important distinctions between traditional SEO and GEO: information gain. In an internet increasingly flooded with generic AI-generated content, the brands that win AI visibility will be the ones that give large language models something genuinely new to learn — proprietary data, original research, first-hand experience, and unique expertise. The GEO Show covers the news, strategies, experiments, and emerging tactics shaping how brands get discovered and cited by ChatGPT, Google AI, Perplexity, Claude, and other AI platforms. If you're responsible for SEO, GEO, content, demand generation, or B2B marketing, subscribe and follow along. 💬 Question: Do you think GEO is fundamentally different from SEO — or is it simply the next evolution of it?

• Transcript

The Anti-Slop Content that Wins AI Love

AI slop isn’t bad because AI helped create it. It’s bad because it adds nothing. Same recycled facts. Same obvious advice. Same five bullet points scraped from the same five sources—rewritten by a machine and published by a company with no genuine expertise in the final output. This is what happens when everyone uses the same models, the same public information and increasingly similar prompts. The result is an internet drowning in synthetic consensus. And publishing more of it won’t win you visibility in Google, ChatGPT, Perplexity, Gemini or any other answer engine. In GEO Day 3, Simon Young and Paris Childress will break down the alternative: anti-slop content. Anti-slop content contains something an AI model cannot produce without access to your company’s knowledge, evidence and experience. It might include: A subject-matter expert’s original point of view First-party research or proprietary data A detailed customer case with specific evidence A tested framework developed through real work An unexpected insight that challenges industry consensus First-hand experience with a problem, product or market A clear, defensible opinion—not a summary of everyone else’s Information that meaningfully advances the existing conversation This is high information gain content: material that leaves the reader—and potentially the large language model (LLM)—knowing something it could not have learned from the other ten articles already ranking for the topic. That distinction matters because Generative Engine Optimization (GEO) is not a content-volume competition. AI systems already have an almost unlimited supply of generic explanations. They do not need your company to publish another “ultimate guide” built from information they can find everywhere else. What they need are useful, credible and sourceable contributions. Content with facts they can extract. Evidence they can reference. Ideas they can attribute. Insights worth citing. During this LinkedIn Live session, we’ll explore: How to recognize AI slop—even when it sounds polished Why “human-written” does not automatically mean high quality What information gain looks like in real B2B content Why proprietary knowledge is the foundation of defensible content How to extract original insight from internal experts How to turn interviews, research and customer evidence into GEO-ready assets Where AI strengthens the process—and where it destroys differentiation How to structure content so its strongest claims are easy to understand, extract and cite How to audit your current content for generic thinking What a scalable anti-slop content workflow looks like This is not an anti-AI session. AI is an extraordinary production and transformation tool. But it cannot manufacture genuine expertise, original evidence or lived experience. If the source material is generic, the output will be generic—regardless of how sophisticated the prompt sounds. The winning model is not AI replacing human expertise. It is human expertise, proprietary knowledge and original evidence—structured and amplified by AI. By the end of this session, you’ll have a clearer standard for deciding what deserves to be published, what belongs in the recycling bin and what has a realistic chance of earning attention, trust and GEO visibility. Because in a world where anyone can generate content, the advantage belongs to companies that actually have something to say. Join Simon and Paris for The Anti-Slop Content That Wins GEO Love.

• Transcript